The eHealth unit in Sierre has been operational since 2011, focusing on applied research and innovation in health technology with a strong collaborative approach. The unit has engaged in multiple projects involving big data to support health decisions, particularly emphasizing early diagnosis and health monitoring. Their work includes developing an infrastructure for managing clinical data and enabling interoperability among various health partners to improve patient care.

1. eHealth unit HES-SO in Sierre
Henning Müller
Michael Schumacher

2. eHealth in Sierre
• History:
– Many eHealth projects since 2007
– eHealth unit since 2011
• Applied research, committed to innovation
• Close to user needs, with strong links:
– Locally (Hôpital VS, Logival, …),
– Nationally (CHUV, HUG, EPFL, …) and
– Internationally (Stanford, Harvard, Imperial
College, Carnegie Mellon, NLM, …)

3. Some of our partners
33333333333333

4. Some numbers
• 22 collaborators
– 3 professors, 5 engineers, 6 postdocs, 8 PhD students
– Many visiting researchers and exchanges with other
research groups & companies
• 60 peer reviewed publications in 2012
• 1 startup company in 2013
• Projects 2013:
– 8 EU FP7 projects
– 4 FNS + 2 Nano-Tera
– CTI, TheArk, Hasler, …
– Mandates

5. Research vision
• Medicine is getting increasingly data intensive
– Digital patient is (becoming) a reality
– Health records, Health monitoring, Internet
information, social networks, genomic data, …
• Our main objective is to support the health
domain
– … by connecting data and people
– … understanding and combining multiple data
sources for reliable interpretations

6. How can we access, use and interpret
data for reliable decision support?
Picture:http://biomedicalcomputationreview.org

8. Interoperability & Semantics
Picture: http://www.teliris.com

9. Picture: http://www.teliris.com

11. Data visualization &
Decision support
Picture: http://www.testroniclabs.com

13. Health monitoring &
expert systems

15. Sustainable Health
Technology
• Aging population & lifestyle (diabetes,
cancer, heart diseases, etc.)
• Need to sustain health to change
behavior & to allow for a healthy living
– Shift focus from treatments to detection
and prevention
– Develop early diagnosis & health
monitoring
• Interdisciplinary:

16. Gestational Diabetes Mellitus
• GDM occurs during pregnancy (4%)
due to increased resistance to insulin
• Goal of the project:
1. Constant monitoring and recording
to ease treatment adjustment
2. Automatic alerts to medical staff
• Technologies:
– Market sensors (glucometers)
– Smart phones & web apps
– Expert systems

18. • VISual Concept Extraction challenge in
RAdioLogy http://visceral.eu/
• EU funded research project on the creation
of a research infrastructure
– Making big image data sets available for
research in image analysis (10-50 TB)

19. Organize 2 competitions
• 1. Extract organs and
landmarks in images
– Map these to semantics
– Allow navigation in data
– Basic task required
• 2. Find similar cases
– Including images and
radiology reports
– Combining images, text
and structured data

20. Our role in VISCERAL
• Create the platform and infrastructure to
manage the research data in the cloud
• Annotate/prepare data
– With radiologists
– Assure interoperability
• Evaluate results
– Assure scalability and automation when
analyzing the data, necessary for big data
• Creation of a gold and silver corpus
– Organize workshops to compare results

21. Why big data in medicine?
• Data production is already enormous and
it will continue to increase (genetics, …)
– Most can not be used for research as this is
private data
• In very large data similar cases can
always be found
– Learn from the past for the future
– Similar in age, anamnesis, co-morbities
– Also for rare diseases that are currently
problematic

22. • Clinically-lead EU project, (Children hospital Rome)
• Follows two past projects, health-e-child and
sim-e-child
• Integrate complex data
and support decisions
• Simulate patients and
outcomes
• Avoid animal testing
• http://www.md-paedigree.eu/

23. Target diseases
• Cardiomyopathies
– Strongly related to imaging
– Simulate treatment outcome
– Personalized care
• Obesity-related cardiovascular
disease
– Strong increase, societal impact
• Juvenile idiopathic arthritis
• Neurological & neuromuscular
diseases

24. Our role
• Creation of an infostructure to manage all
clinical & research data in the project
– Assure semantic interoperability between the
different clinical partners
– Integrate the data
• Support physicians to find “patients like
mine” and patients to find “patients like me”
– Use structured data, free text and imaging data
combined for similar case retrieval
– Currently analyzing the requirements

25. Conclusions
• The digital patient is a reality
– Increasingly complex data in large amounts
• Collaboration between all partners in the
health system is required
– Management of big data and use of extracted
information for decision making
• Many technical challenges
– Temporal data, images, semantics
• Sustainable health is the goal of research

26. More on our research
• Contact:
– Henning.Mueller@hevs.ch
– Michael.Schumacher@hevs.ch
• More information:
– http://publications.hevs.ch/
– http://medgift.hevs.ch/
– http://aislab.hevs.ch/